DESCRIPTION: The primary visual cortex has been thought of as a set of independent, static, quasi-linear visual filters. Evidence now suggests that each cell is a node in a dynamic nonlinear network that has the potential for high-level tasks such as figure-ground discrimination and feature integration. The proposed work is intended to bridge the gap between the single-cell;l and network model of visual cortex by exploring cell behavior in the context of activity across groups of cells. This will be done in three ways: (1) modified reverse correlation and dual stimulation paradigms will be used to create a high-resolution map of the influences of peripheral stimulation on Classical Receptive Field (CRF) response characteristics. Anatomical labels and tracers will be used to determine both the lateral and layer organization of excitatory and inhibitory peripheral pathways. (2) Spike-train oscillation may be a means of identifying salient stimuli. Spike bursting is a viable mechanism for enhancing information transfer between cells. The links between oscillation, bursting and cell coupling will be studied by simultaneously recording activity across groups of 2-6 cells and assessing the relationship between oscillation and bursting, as well as how these mechanisms support synaptic communication. The hypothesis that oscillation represents stimulus salience will be tested by analysis of the modulation of oscillation and connection strength by different stimuli. (3) In addition to the information in a spike train, visual stimuli might be represented by dynamic sub-groups of cells. Simultaneous activity in groups of cells will be observed to see how correlation in their responses changes as a function of different spatial configurations of the stimulus. Responses will be evaluated using type analysis, which detects ensemble coding and determines whether it is based on spike rate or synchronization. The question of whether non-linear influences on the cortical response are calculated locally (unique to the cell) or globally (common to the group) will be examined by seeing if different stimuli cause common changes in synaptic effectiveness throughout a group. This work is fundamental to understanding the distributed representation of sensory information in the brain.